Abe Korah and Erin Dorris Cassidy
Print version (Adobe Reader required)
This study assessed student use of and satisfaction with the WebFeat federated search tool, which was implemented by the library at Sam Houston State University. Students voluntarily responded to an electronic survey, providing feedback on how often they conducted class research using the federated search tool, individual databases, and online search engines and how well each search tool satisfied their class research needs. The study found a high rate of federated search use but only moderate satisfaction; for most students, federated search did not replace individual databases and online search engines, which also saw frequent use for class assignments. Federated search use was highest among lower-level undergraduates, and both use and satisfaction declined as student classification rose. Classification—which can be seen as the amount of experience in an academic environment—played a larger role in federated search use and satisfaction than did age or subject area of study. Students have almost unlimited avenues through which to gather information for conducting research, both in libraries and online. Recent years have seen an increase in the quantity and popularity of free Web-based resources, such as Wikipedia. Regardless of the comparable quality of data, these tools present information in a simple, user-friendly way and require little formal knowledge of information organization and searching techniques. Such straightforward simplicity attracts many students, and academic libraries face challenges in capturing and keeping students’ attention to assist them in finding authoritative and appropriate research materials in the library.
Federated search systems—alternatively called metasearch systems—aim to search a collection of databases from one interface and present one set of results, thereby reducing the amount of time and energy that a researcher must invest in learning and using individual database interfaces.
Although federated search systems are, conceptually, an ideal way to simplify the search process, in practice they often suffer from certain weaknesses, including slowness, fewer advanced search refinements, and poor integration of results from multiple sources. Many problems stem primarily from a lack of consistency between database systems. However, despite such common weaknesses, federated search systems can provide a relatively quick and simple mechanism for conducting a broad search of multiple resources in one step.
In spring 2007, several teams of students in John Newbold’s class in strategic marketing management at Sam Houston State University (SHSU) were given the assignment of producing a marketing plan for the university’s Newton Gresham Library. Some of the teams surveyed students on campus, asking how the library could better market its online resources, while other teams relied on their own preferences and suggestions.
The opinions from the teams and survey respondents showed a desire for a more Google-like approach to searching library resources; students were accustomed to using Google and other Internet search engines to search once and retrieve a single, simple list of results from many websites, ranked by relevancy. That familiarity created the expectation that the library should provide a similar capability for quick, convenient academic research. In response to this finding, the Newton Gresham Library researched metasearch options and finally implemented the federated search product WebFeat, which was marketed on the library’s website under the name E-Z Search.
The federated search tool was released in a beta version on the library website in August 2007. E-Z Search was marketed through the library website, library instruction sessions, and handouts available at the library reference desk. In addition to the new E-Z Search tool, students still had access to the library’s online catalog (branded Sam-Cat) and the native search interfaces for approximately 180 subscription databases. After about six months of use, the library collected information about how many users were searching with E-Z Search and whether it was satisfying their academic search needs. The library conducted the inquiry through an electronic survey, which was designed to answer the following questions:
- Which students are using E-Z Search? How are they using it, and how often?
- How do students perceive E-Z Search, and how well does it satisfy their academic search needs?
- In student opinions, how does E-Z Search compare to other library search tools (the online catalog and individual database interfaces) and Internet search engines?
This article highlights the Newton Gresham Library’s findings concerning the use and perception of the E-Z Search federated search implementation.
Much of the literature on federated search discusses creating and implementing federated search tools and compares various tools and usability studies.1 At the time this article was written, there was not a large pool of quantitative data about user desires and satisfaction with federated searching in an academic environment.
Students have multiple tools at their disposal when conducting research for academic purposes, including their library’s catalog and databases as well as websites and Internet search engines. The Electronic Publishing Initiative at Columbia (EPIC) online survey concluded that almost 50 percent of students started searches for class assignments using a commercial search engine.2 Jillian R. Griffith’s research found that “45 percent of students use Google as their first port of call when locating information, with the university library catalogue used by 10 percent of the sample.”3 Helen Laurence and William Miller believe that this is because “library patrons expect to find it all in cyberspace …, but for the purposes of academic research, such expectations are unrealistic and even dangerous.”4
Libraries have tried adapting to the expectations of users by providing a single search box interface that mimics popular Internet search interfaces. Morgan asserts that commercial web-sites’ characteristics, such as aesthetics and navigation, are the benchmarks that patrons use to judge the viability of a federated search interface.5 Students want a simple interface, and they have no desire to read instructions before starting a search.6 For instance, Google does provide advanced tools for more experienced searchers, but an individual with only average or limited search experience can begin searching with Google almost instantaneously; it requires virtually no instruction to begin searching and interpreting results. In fact, Jung et al. concluded that providing interfaces similar to commercial search engines is crucial to getting undergraduates to use federated searches because their familiarity will increase their confidence in starting their search. Users also expect relevance, speed, and spell-check functions that are comparable to popular search engines.7
Pages: 1 2 3 4